US5996048A - Inclusion vector architecture for a level two cache - Google Patents

Inclusion vector architecture for a level two cache Download PDF

Info

Publication number
US5996048A
US5996048A US08/879,530 US87953097A US5996048A US 5996048 A US5996048 A US 5996048A US 87953097 A US87953097 A US 87953097A US 5996048 A US5996048 A US 5996048A
Authority
US
United States
Prior art keywords
cache
level
line
level cache
inclusion
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
US08/879,530
Inventor
Rajasekhar Cherabuddi
Ricky C. Hetherington
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Oracle America Inc
Original Assignee
Sun Microsystems Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sun Microsystems Inc filed Critical Sun Microsystems Inc
Priority to US08/879,530 priority Critical patent/US5996048A/en
Assigned to SUN MICROSYSTEMS, INC. reassignment SUN MICROSYSTEMS, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: CHERABUDDI, RAJASEKHAR, HETHERINGTON, RICKY C.
Application granted granted Critical
Publication of US5996048A publication Critical patent/US5996048A/en
Assigned to Oracle America, Inc. reassignment Oracle America, Inc. MERGER AND CHANGE OF NAME (SEE DOCUMENT FOR DETAILS). Assignors: Oracle America, Inc., ORACLE USA, INC., SUN MICROSYSTEMS, INC.
Anticipated expiration legal-status Critical
Application status is Expired - Lifetime legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02Addressing or allocation; Relocation
    • G06F12/08Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0806Multiuser, multiprocessor or multiprocessing cache systems
    • G06F12/0811Multiuser, multiprocessor or multiprocessing cache systems with multilevel cache hierarchies

Abstract

A cache architecture with a first level cache and a second level cache, with the second level cache lines including an inclusion vector which indicates which portion of that line are stored in the first level cache. In addition, an instruction/data bit in the inclusion vector indicates whether a portion of that line is in the instruction cache at all. Thus, when a snoop is done to the level two cache, additional snoops to the level one cache only need to be done for those lines which are indicated as present by the inclusion vector.

Description

BACKGROUND OF THE INVENTION

The present invention relates to microprocessors with two levels of cache, and in particular to those with a level two cache on-chip with a larger line size than the level one cache, and having inclusion.

With the high on-chip transistor integration technologies available today, modern microprocessor architectures are able to incorporate a small level one cache, typically broken up into a level one instruction cache and a level one data cache. This is typically backed up with a large level two cache which contains both instructions and data. Typically, the line size of a level one cache is much shorter, and is dictated by the number of address and data lines which can be internally routed on the chip and the machine architecture that demands tremendous level two cache bandwidth and minimum latency. In the level two cache, on the other hand, the line size is more dependent upon the external bus size or the size of a connection to a level three dedicated cache. An ideal line size for a level two cache might be up to four times or more longer than the level one cache line size for the optimal design conditions, considering the system bandwidth and the internal level one and level two cache bandwidth.

However, such a ratio of the level two to level one line size causes a problem in a multi-processor architecture where there is snooping on the level two cache. Snooping is done to maintain data consistency where multiple processors are operating on the same data. While a single snoop may go to a single line in a level two cache, this then requires multiple snoops to the level one cache. Even with a dual ported level one cache, such a number of snoops can interfere with the processor operation. Accordingly, typically designs keep the ratio of the level two line size to the level one line size to be two to one at the most.

It would be desirable to have a higher ratio of level two to level one cache line size, while reducing the number of snoops required.

SUMMARY OF THE INVENTION

The present invention provides a cache architecture with a first level cache and a second level cache, with the second level cache lines including an inclusion vector which indicates which portions of that line are stored in the first level cache. In addition, an instruction/data bit in the inclusion vector indicates whether a portion of that line is in the instruction cache at all.

Thus, when a snoop is done to the level two cache, additional snoops to the level one cache only need to be done for those lines which are indicated as present by the inclusion vector. The present invention also limits the number of level one cache accesses for other than snoop operations. For example, when a level two cache line is victimized, each corresponding level one cache line would normally need to be invalidated. The inclusion vector enables a determination of which level one lines need to be invalidated, without requiring an inquiry of all the level one cache lines.

Similarly, for self-modifying code, data and then an instruction is typically written to the level two cache. The inclusion vector allows a determination of whether there is a need to write the instruction to the instruction cache or invalidate it.

Although not necessary, the present invention is preferably used in a cache system using inclusion. Inclusion means that every entry in a level one cache is also in the level two cache.

The present invention also allows synonym detection. Where a cache architecture has a virtual first level cache, and a physical second level cache, there can be a miss in the first level cache and a hit in the second level cache if the physical address in the second level cache corresponds to a different virtual address than what is stored in the first level cache. In this instance, a first level cache virtual address may be a synonym of the virtual address which would cause a hit, since both relate to the same physical address. In this case, the presence of the synonym can be detected if the inclusion vector is set, allowing it to be invalidated in the level one cache.

The inclusion vector of the present invention preferably also includes a dirty bit. This dirty bit will be set when a portion of the line in the level two cache has been modified in the level one cache. This allows a reduction in snoop response latency. Upon a snoop, this bit in the second level cache can be checked to see if there is more modified data in the first level cache. This reduces the need to always check the first level cache where the dirty bit is not set.

For a further understanding of the nature and advantages of the invention, reference should be made to the following description taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of a typical prior art multi-processor system including three levels of caches.

FIG. 2 illustrates the level one and level two cache structures of the present invention, including the inclusion vector.

FIG. 3 is a timing diagram illustrating the latency of snoop operations.

FIGS. 4A-4C are flowcharts illustrating the use of an inclusion vector in a snoop operation according to the invention.

FIG. 5 is a flowchart illustrating the use of the inclusion vector upon a victimization.

FIG. 6 is a flowchart illustrating the use of an inclusion vector with self-modifying code.

FIG. 7 is a flowchart illustrating the use of the inclusion vector for synonym detection.

DESCRIPTION OF THE PREFERRED EMBODIMENT

FIG. 1 illustrates an example of a multi-processor architecture. Shown in FIG. 1 are three microprocessors 12, 14, and 16 connected to a bus 18. Additional microprocessors could be connected, or fewer microprocessors could be connected. Each microprocessor include its own, dedicated level three cache, labelled 20, 22, and 24. Main memory 26 is also connected to the bus, as is a snoop controller 28.

Internally, at a high level, microprocessor 12 includes a level one instruction cache 30 and a level one data cache 32. A level two cache 34 that is much larger is also included. In very general terms, the microprocessor operates with instructions from instruction cache 30 which are fed to an instruction decoder 36, and then to an instruction execution unit or pipeline 38. The instruction execution will typically access data in a register file 40. Additional data may be retrieved from the level one cache. If data or instructions are not available in the level one instruction or data caches 30 and 32, reference can then be made to the level two cache 34. Upon a miss in level two cache 34, level three cache 20 could be accessed, or main memory 26.

In one embodiment, the level one caches can have a line size of 32 bytes, while the level two cache would have a line size of 128 bytes, or have each line contain four lines of an level one cache. The line size of the level one cache is limited by the ability to route the buses internally. The level two cache will often have a larger line size so that it can interface with a larger, dedicated interface, level three cache 20 or to system bus 18.

When a particular microprocessor needs data, modifies data or does other operations on data, other microprocessors need to be apprised or searched as appropriate, since the data is shared. For example, microprocessor 14 may want the most recent copy of a particular piece of data at a particular address. It will issue a snoop request to snoop controller 28, which will then snoop on the internal caches of the other microprocessors to determine if it is present in any of those, and has been recently modified or updated.

A number of snooping protocols are provided, which typically store bits with each line in the cache to indicate its status. For example the MOESI protocol stores bits indicating whether the cache line has been modified, is exclusive, is shared, or has been invalidated. A number of other snooping protocols are also possible. As can be seen, a snoop to level two cache 34 for a particular line requires only a single line to be accessed. However, to prorogate that snoop to the level one caches requires four snoops internally, since the line size of the level two cache covers four lines of the level one cache. The present invention provides a mechanism for cutting down this number of internal snoops.

FIG. 2 illustrates the data structure for the level one and level two cache according to the present invention. For illustration, the level one (L1) data cache is shown, having a number of lines. Each group of lines is divided into a number of quadrants 0-3 in a data array 42. The address is stored in the corresponding tag array 44, and is preferably a virtual address to allow quick access without translation. Level two (L2) cache 34 includes a data array 46. Each cache line contains data which would be in four lines in the level one cache. A corresponding tag array 48 includes the physical tag corresponding to the line. In addition, an inclusion vector 50 according to the present invention is provided.

The inclusion vector includes four bits 52. Each of bits 52 corresponds to a particular quadrant in the level one cache. Setting the bit to a one indicates that data for that quadrant is present in the level one cache. Thus, for example, if bits 52 are 0100, only quadrant one needs to accessed, since the remaining quadrants 0, 2 and 3 in the level one cache do not include data from that line in the level two cache. Although four bits are shown, and four quadrants, any other number could be used, such as two bits for two quadrants, six bits for six quadrants, eight bits for eight quadrants, etc.

In addition, inclusion vector 50 preferably includes an instruction/data bit 54. This bit can indicate whether the information in that line is in the instruction cache or the data cache, or can simply indicate whether or not it is present in the instruction cache. The other four bits 52 will serve the purpose of indicating which portions of the data are in the level one data or instruction cache.

Finally, a dirty bit 56 is included in the inclusion vector. This bit will indicate whether any of the data in the level one cache has been more recently modified than data in the level two cache. This is useful for reducing latency in the GET portion of a snoop, as will be discussed in more detail below.

FIG. 3 illustrates the timing of snoop in a system such as shown in FIG. 1. First, a active data snoop (ADS) signal 60 is activated, with the corresponding snoop address 62. This will start a snoop check within each microprocessor, with a snoop response-check signal 64 being provided several cycles later to indicate whether the data is present or not. Subsequently, a snoop response-data signal 66 indicates that the data will be provided, if it is present, or if it has been modified, dependent upon the type of snoop request. Subsequently, a data ready signal 68 is provided, along with the actual data 70. Most systems have a fixed or deterministic latency for the snoop response signal, and this invention allows that latency to be met.

FIGS. 4A-4C are flow charts illustrating a snoop transaction. Referring to FIG. 4A, a snoop transaction is first initiated (step 72) and a test is done to see if it has been aborted by the microprocessor (step 74). If it has not been aborted, a snoop check routine (step 76) is run by each microprocessor, with the details shown in FIG. 4B. Then, the MOESI bits are examined to determine if the snooped line is in the M (modified) state (step 78). If it is modified, a snoop GET routine is run (step 80), is shown in more detail in FIG. 4C.

FIG. 4B illustrates a snoop check routine 76. First, the level two cache is examined to determine if the data is present (step 82). If it is not, the snoop check is completed and an appropriate signal can be returned. If it is present, the line is then tested to see if it is in the modified state (step 84). If it is, the snoop check routine is completed, and the snoop GET will then be done as shown in FIG. 4C. If it is not, a tag write is then done to the level two cache in accordance with the MOESI protocol. The inclusion vector is then examined to determine whether the data is in the level one cache (step 88). If it is, a snoop to the appropriate level one cache location is generated (step 90). A tag write in accordance with the MOESI or other snoop protocol is then done at the level one cache (step 92).

FIG. 4C shows the snoop GET routine 80. The level one and level two caches are checked for the presence of the required data (step 94). If the data is present, a read from the appropriate cache is done (step 96). Subsequently, a tag write in accordance with the MOESI or other snoop protocol is done in the level two cache (step 98) and, if appropriate, the level one cache (step 100). As can be seen, the use of the inclusion vector dramatically reduces the number of snoops to the level one cache which are required. The number is reduced to only that number that is absolutely necessary to accomplish the snoop.

The inclusion vector according to the present invention also reduces accesses to the level one cache for other types of operations. For example, as shown in FIG. 5, the victimization of a line in the level two cache can utilize the inclusion vector. A victimization occurs when a cache line needs to be written over in the level two cache for new data pulled from main memory. An algorithm such as the least recently used (LRU) algorithm may be used to determine which line to "victimize", or write over. When that line is victimized, it needs to be determined whether any of the data in that line is also in the level one caches. Typically, this would require four accesses to each of the four corresponding lines in the level one cache where the data might be located, to determine if it is present, and to invalidate the data in the level one cache. The present invention limits the number of accesses required.

As shown in the flow chart of FIG. 5, a fetch request for the new data into a line of the level two cache is first performed (step 102). The data in the line to be written over is examined to determine if it is in the modified state (step 104), and if it is, the data to be written over is first saved by writing it to the level three cache or to main memory (step 106). The inclusion vector is then checked (step 108) to determine if any of the data is also in the level one cache. To the extent that it is present, only the level one cache lines indicated by the inclusion vector are invalidated (step 110). After this has been accomplished, new data can be written over the line in the L2 cache (step 112). And, if appropriate in accordance with a write-through or other protocol, the portion of the new line required by the microprocessor can be written into the level one cache (step 114). Finally, the new inclusion vector can be written (step 116) either indicating which portions have been written in the level one cache, or indicating all zeros if it is not written into the level one cache.

As shown by the flow chart of FIG. 6, the present invention can also be used when self-modifying code modifies a location in the on-chip cache. Self-modifying code will typically write new data, and then subsequently modify an instruction which either uses the new data or modifies the instruction in accordance with the new data. Thus, as shown in FIG. 6, first the new data is written to the level two cache (step 118). The inclusion vector is then examined to determine if any of that data that is modified is also in the level one cache (step 120). If it is, the data is written through to the level one cache location, or the level one cache lines are invalidated (step 122). Next, an instruction can be written to the level two cache (step 124). The inclusion vector can then be examined, with the single instruction/data bit being looked at, to determine if that instruction is in the level one instruction cache (step 126). If it is, again, the instruction can be written through to the level one cache, or invalidated there (step 128).

The inclusion vector of the present invention also has use for synonym detection. A synonym is where there are two or more virtual addresses corresponding to the same physical address. In an architecture where the level one cache is a virtual cache to allow quick access without requiring translation, and the level two cache is a physical cache, there is a need to detect synonyms.

As illustrated in FIG. 7, this occurs upon a read transaction (step 130) being performed to fetch data. If there is a hit in the level one cache (step 132) there is no need for the synonym detection. However, if there is no hit, a translation of the virtual addresses then done to a physical address (step 134). If there is no hit in the level two cache (step 136), again no synonym detection needs to be done, since without the physical address data being there, none of the synonyms could be present in the level one cache in a system using inclusion.

However, if there is a hit in the level two cache, there is a need to determine whether there is a synonym present in the level one cache. This is done by examining the inclusion vector to see if it is set (step 138). If it is not set, there is no synonym present. If it is set, there is a synonym present, and that synonym can be invalidated in the level one cache (step 140), and need only be invalidated for the particular quadrants present as indicated by the inclusion vector, thus providing a double savings in time. The data can then be provided, or could be provided concurrently, from the level two cache to the execution unit of the microprocessor (step 142).

As will be understood by those skilled in the art, the present invention my be embodied in other specific forms without departing from the spirit or essential characteristics thereof. Accordingly, the foregoing description is intended to illustrative, but not limiting, of the scope of the invention which is set forth in the following claims.

Claims (16)

What is claimed is:
1. A cache architecture comprising:
a first level cache having a first line length;
a second level cache having a second line length longer than said first line length; and
an inclusion vector for each line of said second level cache, said inclusion vector comprising a plurality of bits indicating which portions of said line of said second level cache are stored in said first level cache.
2. The cache architecture of claim 1 wherein said inclusion vector includes a separate bit for each line of said first level cache corresponding to a portion of a line of said second level cache.
3. The cache architecture of claim 1 further comprising:
a first level instruction cache;
a first level data cache; and
an instruction cache inclusion bit stored in said second level cache for each of a plurality of lines of said second level cache, said instruction cache inclusion bit being a separate bit from said inclusion vector bits and indicating whether any portion of an associated second level cache line is stored in said instruction cache at all.
4. The cache architecture of claim 1 wherein said inclusion vector is stored in a tag array of said second level cache.
5. The cache architecture of claim 1 wherein said first level cache is a virtual tag cache, and said second level cache is a physical tag cache.
6. The cache architecture of claim 1 wherein said inclusion vector further comprises a dirty bit for each of at least a plurality of lines, said dirty bit indicating whether a portion of said line has been modified in said first level cache.
7. A cache architecture comprising:
a first level instruction cache;
a first level data cache having a first line length;
a second level cache having a second line length longer than said first line length; and
an inclusion vector for each line of said second level cache, said inclusion vector comprising a plurality of bits indicating which portions of said line of said second level cache are stored in said first level cache, said inclusion vector including a separate bit for each line of said first level cache corresponding to a portion of a line of said second level cache; and
an instruction cache inclusion bit stored in said second level cache for each of a plurality of lines of said second level cache, said instruction cache inclusion bit being a separate bit from said inclusion vector bits and indicating whether any portion of an associated second level cache line is stored in said instruction cache at all.
8. The cache architecture of claim 7 wherein said inclusion vector is stored in a tag array of said second level cache.
9. The cache architecture of claim 7 wherein said first level cache is a virtual tag cache, and said second level cache is a physical tag cache.
10. The cache architecture of claim 7 wherein said inclusion vector further comprises a dirty bit for each of at least a plurality of lines, said dirty bit indicating whether a portion of said line has been modified in said first level cache.
11. A computer system comprising:
a system bus;
a main memory coupled to said bus; and
a plurality of processors coupled to said bus, at least one of said processors having a cache architecture including
a first level cache having a first line length,
a second level cache having a second line length longer than said first line length, and
an inclusion vector for each line of said second level cache, said inclusion vector comprising a plurality of bits indicating which portions of said line of said second level cache are stored in said first level cache.
12. A method for constructing a cache architecture comprising the steps of:
providing a first level cache having a first line length;
providing a second level cache having a second line length longer than said first line length; and
providing an inclusion vector for each line of said second level cache, said inclusion vector comprising a plurality of bits indicating which portions of said line of said second level cache are stored in said first level cache.
13. A method for operating a cache architecture having a first level cache having a first line length and a second level cache having a second line length longer than said first line length, comprising the steps of:
providing an inclusion vector for each line of said second level cache, said inclusion vector comprising a plurality of bits indicating which portions of said line of said second level cache are stored in said first level cache; and
examining said inclusion vector before accessing said first level cache to determine which portions of said first level cache need to be accessed.
14. The method of claim 13 wherein said accessing is for writing through data to said first level cache.
15. The method of claim 13 wherein said accessing is for invalidating any relevant line in said first level cache as a result of a victimization in said second level cache.
16. The method of claim 13 wherein said first level cache is a virtual cache, said second level cache is a physical cache, and further comprising the steps of:
examining said inclusion vector upon a miss in said level one cache and a hit in said level two cache;
invalidating any indicated entries in said level one cache, said indicated entries necessarily being synonym entries.
US08/879,530 1997-06-20 1997-06-20 Inclusion vector architecture for a level two cache Expired - Lifetime US5996048A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US08/879,530 US5996048A (en) 1997-06-20 1997-06-20 Inclusion vector architecture for a level two cache

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US08/879,530 US5996048A (en) 1997-06-20 1997-06-20 Inclusion vector architecture for a level two cache

Publications (1)

Publication Number Publication Date
US5996048A true US5996048A (en) 1999-11-30

Family

ID=25374337

Family Applications (1)

Application Number Title Priority Date Filing Date
US08/879,530 Expired - Lifetime US5996048A (en) 1997-06-20 1997-06-20 Inclusion vector architecture for a level two cache

Country Status (1)

Country Link
US (1) US5996048A (en)

Cited By (47)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6345339B1 (en) * 1998-02-17 2002-02-05 International Business Machines Corporation Pseudo precise I-cache inclusivity for vertical caches
US6360301B1 (en) * 1999-04-13 2002-03-19 Hewlett-Packard Company Coherency protocol for computer cache
US6490660B1 (en) * 1997-08-06 2002-12-03 International Business Machines Corporation Method and apparatus for a configurable multiple level cache with coherency in a multiprocessor system
US6526496B1 (en) * 1999-10-21 2003-02-25 International Business Machines Corporation Burst instruction alignment method apparatus and method therefor
US6574710B1 (en) * 2000-07-31 2003-06-03 Hewlett-Packard Development Company, L.P. Computer cache system with deferred invalidation
US6587923B1 (en) * 2000-05-22 2003-07-01 International Business Machines Corporation Dual line size cache directory
US20040103251A1 (en) * 2002-11-26 2004-05-27 Mitchell Alsup Microprocessor including a first level cache and a second level cache having different cache line sizes
US20040148473A1 (en) * 2003-01-27 2004-07-29 Hughes William A. Method and apparatus for injecting write data into a cache
US20040215901A1 (en) * 2000-12-08 2004-10-28 Gaither Blaine Douglas Method and mechanism to use a cache to translate from a virtual bus to a physical bus
US20050102473A1 (en) * 2002-11-21 2005-05-12 Fujitsu Limited Cache control method and processor system
US20060075193A1 (en) * 2002-05-06 2006-04-06 Sony Computer Entertainment Inc. Methods and apparatus for controlling hierarchical cache memory
US20060143408A1 (en) * 2004-12-29 2006-06-29 Sistla Krishnakanth V Efficient usage of last level caches in a MCMP system using application level configuration
US20060143404A1 (en) * 2004-12-27 2006-06-29 Yen-Kuang Chen System and method for cache coherency in a cache with different cache location lengths
US7133975B1 (en) * 2003-01-21 2006-11-07 Advanced Micro Devices, Inc. Cache memory system including a cache memory employing a tag including associated touch bits
US7334102B1 (en) 2003-05-09 2008-02-19 Advanced Micro Devices, Inc. Apparatus and method for balanced spinlock support in NUMA systems
US20100281235A1 (en) * 2007-11-17 2010-11-04 Martin Vorbach Reconfigurable floating-point and bit-level data processing unit
US7928763B2 (en) 2002-09-06 2011-04-19 Martin Vorbach Multi-core processing system
US20110161977A1 (en) * 2002-03-21 2011-06-30 Martin Vorbach Method and device for data processing
US7996827B2 (en) 2001-08-16 2011-08-09 Martin Vorbach Method for the translation of programs for reconfigurable architectures
US8058899B2 (en) 2000-10-06 2011-11-15 Martin Vorbach Logic cell array and bus system
US8069373B2 (en) 2001-09-03 2011-11-29 Martin Vorbach Method for debugging reconfigurable architectures
US20110320720A1 (en) * 2010-06-23 2011-12-29 International Business Machines Corporation Cache Line Replacement In A Symmetric Multiprocessing Computer
US8099618B2 (en) 2001-03-05 2012-01-17 Martin Vorbach Methods and devices for treating and processing data
US8127061B2 (en) 2002-02-18 2012-02-28 Martin Vorbach Bus systems and reconfiguration methods
US8145881B2 (en) 2001-03-05 2012-03-27 Martin Vorbach Data processing device and method
US8156284B2 (en) 2002-08-07 2012-04-10 Martin Vorbach Data processing method and device
US8195856B2 (en) * 1996-12-20 2012-06-05 Martin Vorbach I/O and memory bus system for DFPS and units with two- or multi-dimensional programmable cell architectures
US8209653B2 (en) 2001-09-03 2012-06-26 Martin Vorbach Router
US8230411B1 (en) 1999-06-10 2012-07-24 Martin Vorbach Method for interleaving a program over a plurality of cells
US8250503B2 (en) 2006-01-18 2012-08-21 Martin Vorbach Hardware definition method including determining whether to implement a function as hardware or software
US8281108B2 (en) 2002-01-19 2012-10-02 Martin Vorbach Reconfigurable general purpose processor having time restricted configurations
US8281265B2 (en) 2002-08-07 2012-10-02 Martin Vorbach Method and device for processing data
US8301872B2 (en) * 2000-06-13 2012-10-30 Martin Vorbach Pipeline configuration protocol and configuration unit communication
US20120302055A1 (en) * 2011-04-22 2012-11-29 Asm Ip Holding B.V. Deposition and reduction of mixed metal oxide thin films
USRE44365E1 (en) 1997-02-08 2013-07-09 Martin Vorbach Method of self-synchronization of configurable elements of a programmable module
US20130346694A1 (en) * 2012-06-25 2013-12-26 Robert Krick Probe filter for shared caches
US8686475B2 (en) 2001-09-19 2014-04-01 Pact Xpp Technologies Ag Reconfigurable elements
US8686549B2 (en) 2001-09-03 2014-04-01 Martin Vorbach Reconfigurable elements
US20140156932A1 (en) * 2012-06-25 2014-06-05 Advanced Micro Devices, Inc. Eliminating fetch cancel for inclusive caches
US8812820B2 (en) 2003-08-28 2014-08-19 Pact Xpp Technologies Ag Data processing device and method
US8819505B2 (en) 1997-12-22 2014-08-26 Pact Xpp Technologies Ag Data processor having disabled cores
US8914590B2 (en) 2002-08-07 2014-12-16 Pact Xpp Technologies Ag Data processing method and device
US20150058571A1 (en) * 2013-08-20 2015-02-26 Apple Inc. Hint values for use with an operand cache
US9037807B2 (en) 2001-03-05 2015-05-19 Pact Xpp Technologies Ag Processor arrangement on a chip including data processing, memory, and interface elements
US9892051B1 (en) * 2008-08-14 2018-02-13 Marvell International Ltd. Method and apparatus for use of a preload instruction to improve efficiency of cache
US10019368B2 (en) 2014-05-29 2018-07-10 Samsung Electronics Co., Ltd. Placement policy for memory hierarchies
US10402344B2 (en) 2013-11-21 2019-09-03 Samsung Electronics Co., Ltd. Systems and methods for direct data access in multi-level cache memory hierarchies

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5119290A (en) * 1987-10-02 1992-06-02 Sun Microsystems, Inc. Alias address support
US5195089A (en) * 1990-12-31 1993-03-16 Sun Microsystems, Inc. Apparatus and method for a synchronous, high speed, packet-switched bus
US5307477A (en) * 1989-12-01 1994-04-26 Mips Computer Systems, Inc. Two-level cache memory system
US5355467A (en) * 1991-06-04 1994-10-11 Intel Corporation Second level cache controller unit and system
US5369753A (en) * 1990-06-15 1994-11-29 Compaq Computer Corporation Method and apparatus for achieving multilevel inclusion in multilevel cache hierarchies
US5440707A (en) * 1992-04-29 1995-08-08 Sun Microsystems, Inc. Instruction and data cache with a shared TLB for split accesses and snooping in the same clock cycle
US5530832A (en) * 1993-10-14 1996-06-25 International Business Machines Corporation System and method for practicing essential inclusion in a multiprocessor and cache hierarchy
US5584013A (en) * 1994-12-09 1996-12-10 International Business Machines Corporation Hierarchical cache arrangement wherein the replacement of an LRU entry in a second level cache is prevented when the cache entry is the only inclusive entry in the first level cache
US5740400A (en) * 1995-06-05 1998-04-14 Advanced Micro Devices Inc. Reducing cache snooping overhead in a multilevel cache system with multiple bus masters and a shared level two cache by using an inclusion field

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5119290A (en) * 1987-10-02 1992-06-02 Sun Microsystems, Inc. Alias address support
US5307477A (en) * 1989-12-01 1994-04-26 Mips Computer Systems, Inc. Two-level cache memory system
US5369753A (en) * 1990-06-15 1994-11-29 Compaq Computer Corporation Method and apparatus for achieving multilevel inclusion in multilevel cache hierarchies
US5195089A (en) * 1990-12-31 1993-03-16 Sun Microsystems, Inc. Apparatus and method for a synchronous, high speed, packet-switched bus
US5355467A (en) * 1991-06-04 1994-10-11 Intel Corporation Second level cache controller unit and system
US5440707A (en) * 1992-04-29 1995-08-08 Sun Microsystems, Inc. Instruction and data cache with a shared TLB for split accesses and snooping in the same clock cycle
US5530832A (en) * 1993-10-14 1996-06-25 International Business Machines Corporation System and method for practicing essential inclusion in a multiprocessor and cache hierarchy
US5584013A (en) * 1994-12-09 1996-12-10 International Business Machines Corporation Hierarchical cache arrangement wherein the replacement of an LRU entry in a second level cache is prevented when the cache entry is the only inclusive entry in the first level cache
US5740400A (en) * 1995-06-05 1998-04-14 Advanced Micro Devices Inc. Reducing cache snooping overhead in a multilevel cache system with multiple bus masters and a shared level two cache by using an inclusion field
US5796980A (en) * 1995-06-05 1998-08-18 Advanced Micro Devices Inc. Reducing cache snooping overhead in a multilevel cache system with inclusion field in shared cache indicating state of data in lower level caches

Cited By (81)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8195856B2 (en) * 1996-12-20 2012-06-05 Martin Vorbach I/O and memory bus system for DFPS and units with two- or multi-dimensional programmable cell architectures
USRE45223E1 (en) 1997-02-08 2014-10-28 Pact Xpp Technologies Ag Method of self-synchronization of configurable elements of a programmable module
USRE44383E1 (en) 1997-02-08 2013-07-16 Martin Vorbach Method of self-synchronization of configurable elements of a programmable module
USRE45109E1 (en) 1997-02-08 2014-09-02 Pact Xpp Technologies Ag Method of self-synchronization of configurable elements of a programmable module
USRE44365E1 (en) 1997-02-08 2013-07-09 Martin Vorbach Method of self-synchronization of configurable elements of a programmable module
US6490660B1 (en) * 1997-08-06 2002-12-03 International Business Machines Corporation Method and apparatus for a configurable multiple level cache with coherency in a multiprocessor system
US8819505B2 (en) 1997-12-22 2014-08-26 Pact Xpp Technologies Ag Data processor having disabled cores
US6345339B1 (en) * 1998-02-17 2002-02-05 International Business Machines Corporation Pseudo precise I-cache inclusivity for vertical caches
US8468329B2 (en) 1999-02-25 2013-06-18 Martin Vorbach Pipeline configuration protocol and configuration unit communication
US6360301B1 (en) * 1999-04-13 2002-03-19 Hewlett-Packard Company Coherency protocol for computer cache
US8312200B2 (en) 1999-06-10 2012-11-13 Martin Vorbach Processor chip including a plurality of cache elements connected to a plurality of processor cores
US8726250B2 (en) 1999-06-10 2014-05-13 Pact Xpp Technologies Ag Configurable logic integrated circuit having a multidimensional structure of configurable elements
US8230411B1 (en) 1999-06-10 2012-07-24 Martin Vorbach Method for interleaving a program over a plurality of cells
US6526496B1 (en) * 1999-10-21 2003-02-25 International Business Machines Corporation Burst instruction alignment method apparatus and method therefor
US6587923B1 (en) * 2000-05-22 2003-07-01 International Business Machines Corporation Dual line size cache directory
US8301872B2 (en) * 2000-06-13 2012-10-30 Martin Vorbach Pipeline configuration protocol and configuration unit communication
US6574710B1 (en) * 2000-07-31 2003-06-03 Hewlett-Packard Development Company, L.P. Computer cache system with deferred invalidation
US8471593B2 (en) 2000-10-06 2013-06-25 Martin Vorbach Logic cell array and bus system
US8058899B2 (en) 2000-10-06 2011-11-15 Martin Vorbach Logic cell array and bus system
US9047440B2 (en) 2000-10-06 2015-06-02 Pact Xpp Technologies Ag Logical cell array and bus system
US7032074B2 (en) * 2000-12-08 2006-04-18 Hewlett-Packard Development Company L.P. Method and mechanism to use a cache to translate from a virtual bus to a physical bus
US20040215901A1 (en) * 2000-12-08 2004-10-28 Gaither Blaine Douglas Method and mechanism to use a cache to translate from a virtual bus to a physical bus
US9075605B2 (en) 2001-03-05 2015-07-07 Pact Xpp Technologies Ag Methods and devices for treating and processing data
US8099618B2 (en) 2001-03-05 2012-01-17 Martin Vorbach Methods and devices for treating and processing data
US9037807B2 (en) 2001-03-05 2015-05-19 Pact Xpp Technologies Ag Processor arrangement on a chip including data processing, memory, and interface elements
US8312301B2 (en) 2001-03-05 2012-11-13 Martin Vorbach Methods and devices for treating and processing data
US8145881B2 (en) 2001-03-05 2012-03-27 Martin Vorbach Data processing device and method
US8869121B2 (en) 2001-08-16 2014-10-21 Pact Xpp Technologies Ag Method for the translation of programs for reconfigurable architectures
US7996827B2 (en) 2001-08-16 2011-08-09 Martin Vorbach Method for the translation of programs for reconfigurable architectures
US8407525B2 (en) 2001-09-03 2013-03-26 Pact Xpp Technologies Ag Method for debugging reconfigurable architectures
US8069373B2 (en) 2001-09-03 2011-11-29 Martin Vorbach Method for debugging reconfigurable architectures
US8429385B2 (en) 2001-09-03 2013-04-23 Martin Vorbach Device including a field having function cells and information providing cells controlled by the function cells
US8686549B2 (en) 2001-09-03 2014-04-01 Martin Vorbach Reconfigurable elements
US8209653B2 (en) 2001-09-03 2012-06-26 Martin Vorbach Router
US8686475B2 (en) 2001-09-19 2014-04-01 Pact Xpp Technologies Ag Reconfigurable elements
US8281108B2 (en) 2002-01-19 2012-10-02 Martin Vorbach Reconfigurable general purpose processor having time restricted configurations
US8127061B2 (en) 2002-02-18 2012-02-28 Martin Vorbach Bus systems and reconfiguration methods
US20110161977A1 (en) * 2002-03-21 2011-06-30 Martin Vorbach Method and device for data processing
US7461207B2 (en) * 2002-05-06 2008-12-02 Sony Computer Entertainment Inc. Methods and apparatus for controlling hierarchical cache memory
US7870340B2 (en) 2002-05-06 2011-01-11 Sony Computer Entertainment Inc. Methods and apparatus for controlling hierarchical cache memory
US20090063772A1 (en) * 2002-05-06 2009-03-05 Sony Computer Entertainment Inc. Methods and apparatus for controlling hierarchical cache memory
US20060075193A1 (en) * 2002-05-06 2006-04-06 Sony Computer Entertainment Inc. Methods and apparatus for controlling hierarchical cache memory
US8281265B2 (en) 2002-08-07 2012-10-02 Martin Vorbach Method and device for processing data
US8156284B2 (en) 2002-08-07 2012-04-10 Martin Vorbach Data processing method and device
US8914590B2 (en) 2002-08-07 2014-12-16 Pact Xpp Technologies Ag Data processing method and device
US7928763B2 (en) 2002-09-06 2011-04-19 Martin Vorbach Multi-core processing system
US8803552B2 (en) 2002-09-06 2014-08-12 Pact Xpp Technologies Ag Reconfigurable sequencer structure
US8310274B2 (en) 2002-09-06 2012-11-13 Martin Vorbach Reconfigurable sequencer structure
US20050102473A1 (en) * 2002-11-21 2005-05-12 Fujitsu Limited Cache control method and processor system
US7330961B2 (en) * 2002-11-21 2008-02-12 Fujitsu Limited Cache control method and processor system
US20040103251A1 (en) * 2002-11-26 2004-05-27 Mitchell Alsup Microprocessor including a first level cache and a second level cache having different cache line sizes
US7133975B1 (en) * 2003-01-21 2006-11-07 Advanced Micro Devices, Inc. Cache memory system including a cache memory employing a tag including associated touch bits
US20040148473A1 (en) * 2003-01-27 2004-07-29 Hughes William A. Method and apparatus for injecting write data into a cache
US7155572B2 (en) * 2003-01-27 2006-12-26 Advanced Micro Devices, Inc. Method and apparatus for injecting write data into a cache
US7334102B1 (en) 2003-05-09 2008-02-19 Advanced Micro Devices, Inc. Apparatus and method for balanced spinlock support in NUMA systems
US8812820B2 (en) 2003-08-28 2014-08-19 Pact Xpp Technologies Ag Data processing device and method
WO2006072064A3 (en) * 2004-12-27 2006-09-14 Intel Corp System and method for cache coherency in a cache with different cache location lengths
US7454576B2 (en) * 2004-12-27 2008-11-18 Intel Corporation System and method for cache coherency in a cache with different cache location lengths
KR100925572B1 (en) * 2004-12-27 2009-11-06 인텔 코포레이션 System and method for cache coherency in a cache with different cache location lengths
CN101088074B (en) 2004-12-27 2010-06-16 英特尔公司 System and method for cache coherency in a cache with different cache location lengths
WO2006072064A2 (en) * 2004-12-27 2006-07-06 Intel Corporation System and method for cache coherency in a cache with different cache location lengths
US20060143404A1 (en) * 2004-12-27 2006-06-29 Yen-Kuang Chen System and method for cache coherency in a cache with different cache location lengths
US7991966B2 (en) * 2004-12-29 2011-08-02 Intel Corporation Efficient usage of last level caches in a MCMP system using application level configuration
US20060143408A1 (en) * 2004-12-29 2006-06-29 Sistla Krishnakanth V Efficient usage of last level caches in a MCMP system using application level configuration
US8250503B2 (en) 2006-01-18 2012-08-21 Martin Vorbach Hardware definition method including determining whether to implement a function as hardware or software
US20100281235A1 (en) * 2007-11-17 2010-11-04 Martin Vorbach Reconfigurable floating-point and bit-level data processing unit
US9892051B1 (en) * 2008-08-14 2018-02-13 Marvell International Ltd. Method and apparatus for use of a preload instruction to improve efficiency of cache
US20110320720A1 (en) * 2010-06-23 2011-12-29 International Business Machines Corporation Cache Line Replacement In A Symmetric Multiprocessing Computer
US8871617B2 (en) * 2011-04-22 2014-10-28 Asm Ip Holding B.V. Deposition and reduction of mixed metal oxide thin films
US20120302055A1 (en) * 2011-04-22 2012-11-29 Asm Ip Holding B.V. Deposition and reduction of mixed metal oxide thin films
US20140156932A1 (en) * 2012-06-25 2014-06-05 Advanced Micro Devices, Inc. Eliminating fetch cancel for inclusive caches
US9058269B2 (en) * 2012-06-25 2015-06-16 Advanced Micro Devices, Inc. Method and apparatus including a probe filter for shared caches utilizing inclusion bits and a victim probe bit
US9122612B2 (en) * 2012-06-25 2015-09-01 Advanced Micro Devices, Inc. Eliminating fetch cancel for inclusive caches
US20130346694A1 (en) * 2012-06-25 2013-12-26 Robert Krick Probe filter for shared caches
US9652233B2 (en) * 2013-08-20 2017-05-16 Apple Inc. Hint values for use with an operand cache
US20150058571A1 (en) * 2013-08-20 2015-02-26 Apple Inc. Hint values for use with an operand cache
US10402344B2 (en) 2013-11-21 2019-09-03 Samsung Electronics Co., Ltd. Systems and methods for direct data access in multi-level cache memory hierarchies
US10409725B2 (en) 2014-05-29 2019-09-10 Samsung Electronics Co., Ltd. Management of shared pipeline resource usage based on level information
US10019368B2 (en) 2014-05-29 2018-07-10 Samsung Electronics Co., Ltd. Placement policy for memory hierarchies
US10031849B2 (en) 2014-05-29 2018-07-24 Samsung Electronics Co., Ltd. Tracking alternative cacheline placement locations in a cache hierarchy
US10402331B2 (en) 2014-05-29 2019-09-03 Samsung Electronics Co., Ltd. Systems and methods for implementing a tag-less shared cache and a larger backing cache

Similar Documents

Publication Publication Date Title
EP0674267B1 (en) Sub-line cache coherent write transactions
US7546422B2 (en) Method and apparatus for the synchronization of distributed caches
US5564035A (en) Exclusive and/or partially inclusive extension cache system and method to minimize swapping therein
US5067078A (en) Cache which provides status information
US5787486A (en) Bus protocol for locked cycle cache hit
JP3431824B2 (en) Access method and processing apparatus
US5119485A (en) Method for data bus snooping in a data processing system by selective concurrent read and invalidate cache operation
US6519685B1 (en) Cache states for multiprocessor cache coherency protocols
US6738836B1 (en) Scalable efficient I/O port protocol
JP3510240B2 (en) Multi-processing system with cache memory
US6636949B2 (en) System for handling coherence protocol races in a scalable shared memory system based on chip multiprocessing
JP2566701B2 (en) Ownership change control device for data unit in the shared cache
US5903911A (en) Cache-based computer system employing memory control circuit and method for write allocation and data prefetch
US5745732A (en) Computer system including system controller with a write buffer and plural read buffers for decoupled busses
US5535352A (en) Access hints for input/output address translation mechanisms
US5774700A (en) Method and apparatus for determining the timing of snoop windows in a pipelined bus
US5155824A (en) System for transferring selected data words between main memory and cache with multiple data words and multiple dirty bits for each address
US6145054A (en) Apparatus and method for handling multiple mergeable misses in a non-blocking cache
US6226713B1 (en) Apparatus and method for queueing structures in a multi-level non-blocking cache subsystem
CN1126046C (en) Cache coherency protocol for data processing system including multi level memory hierarchy
US5359723A (en) Cache memory hierarchy having a large write through first level that allocates for CPU read misses only and a small write back second level that allocates for CPU write misses only
US6901495B2 (en) Cache memory system allowing concurrent reads and writes to cache lines to increase snoop bandwith
US20030065884A1 (en) Hiding refresh of memory and refresh-hidden memory
EP0667578B1 (en) Apparatus and method for checking cache coherency with double snoop mechanism
US5796980A (en) Reducing cache snooping overhead in a multilevel cache system with inclusion field in shared cache indicating state of data in lower level caches

Legal Events

Date Code Title Description
AS Assignment

Owner name: SUN MICROSYSTEMS, INC., CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:CHERABUDDI, RAJASEKHAR;HETHERINGTON, RICKY C.;REEL/FRAME:009068/0632

Effective date: 19980109

STCF Information on status: patent grant

Free format text: PATENTED CASE

FPAY Fee payment

Year of fee payment: 4

REMI Maintenance fee reminder mailed
FPAY Fee payment

Year of fee payment: 8

FPAY Fee payment

Year of fee payment: 12

AS Assignment

Owner name: ORACLE AMERICA, INC., CALIFORNIA

Free format text: MERGER AND CHANGE OF NAME;ASSIGNORS:ORACLE USA, INC.;SUN MICROSYSTEMS, INC.;ORACLE AMERICA, INC.;REEL/FRAME:037270/0742

Effective date: 20100212